Towards transferability in fish migration models: A generic operational tool for predicting silver eel migration in rivers

• Published in: Science of The Total Environment
• Main Authors: Teichert, Nils, Tetard, Stéphane, Trancart, Thomas, de Oliveira, Eric, Acou, Anthony, Carpentier, Alexandre, Bourillon, B., Feunteun, Eric
• Other Authors: Biologie des Organismes et Ecosystèmes Aquatiques (BOREA), Université de Caen Normandie (UNICAEN), Normandie Université (NU)-Normandie Université (NU)-Muséum national d'Histoire naturelle (MNHN)-Institut de Recherche pour le Développement (IRD)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Université des Antilles (UA), Centre De Recherche et d'Enseignement sur les Systèmes Côtiers (CRESCO), Muséum national d'Histoire naturelle (MNHN)-Institut Français de Recherche pour l'Exploitation de la Mer (IFREMER), Laboratoire National d’Hydraulique et Environnement (EDF R&D LNHE), EDF R&D (EDF R&D), EDF (EDF)-EDF (EDF), Patrimoine naturel (PatriNat), Muséum national d'Histoire naturelle (MNHN)-Centre National de la Recherche Scientifique (CNRS)-Office français de la biodiversité (OFB), Pôle OFB-INRAE-Institut Agro-UPPA pour la gestion des migrateurs amphihalins dans leur environnement (MIAME), Université de Pau et des Pays de l'Adour (UPPA)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Office français de la biodiversité (OFB)-Institut Agro Rennes Angers, Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro)-Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement (Institut Agro), Université de Rennes (UR), European Commission, EC: LIFE16 NAT/BE/000807Muséum National d'Histoire Naturelle, MNHN
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2020
• Subjects: Eel runs prediction; Ensemble modelling; European eel management; Fish migration; Model transferability; River obstacle; [SDE]Environmental Sciences; [SDE.BE]Environmental Sciences/Biodiversity and Ecology; [SDV.BA.ZV]Life Sciences [q-bio]/Animal biology/Vertebrate Zoology; Anguilla anguilla
• Online Access: https://hal.science/hal-02932307; https://hal.science/hal-02932307/document; https://hal.science/hal-02932307/file/Teichert%20et%20al-2020-Towards%20transferability%20in%20fish%20migration%20models.pdf; https://doi.org/10.1016/j.scitotenv.2020.140069

International audience In the global context of river fragmentation, predicting fish migration is urgent to implement management actions aimed at protecting and promoting the free movement of diadromous fish. However, large-scale applicability of conservation measures requires transferable models that enable prediction of migration even in data-poor regions. Here, we surveyed 12 contrasted European river sites to predict the activity peaks of silver eels (Anguilla anguilla) during river migration towards spawning areas through an ensemble modelling approach. Site-specific Boosted Regression Tree (BRT) models were adjusted using standardized hydrological variables to predict migration probability, which were aggregated in consensus predictions. Results of independent cross-validations demonstrated that silver eel migration runs were accurately predicted in response to changes in river discharge. Transferability and predictive performance were improved by considering catchment-size dissimilarity between river sites (85 to 109,930 km2) when combining the site-specific predictions. Nevertheless, we provided two examples for which the effects of human actions on flow conditions were so high that they prevented reliable predictions of migration runs. Further contributions should thus take advantage of the flexibility of our approach for updating model collection with new sites to extend the predictive performance under a larger range of ecological conditions. Our transferable hydrological-based modelling framework offers an opportunity to implement large-scale management strategies for eel conservation, even in rivers where eel monitoring data lack. The BRT models and prediction functions were compiled in an R package named ‘silvRpeak’ to facilitate operational implementation by end-user managers, which can determine when mitigation measures should be implemented to improve river continuity (e.g. turbine shutdown and sluice gate opening) and balance their economic activity towards eel conservation. The only input ...